Outer retinal corrugations in late-onset retinal degeneration: a diagnostic finding demonstrated with multimodal imaging

Introduction

Late-onset retinal degeneration (L-ORD) is a rare autosomal dominant inherited retinal degeneration caused by a mutation in the C1QTNF5 gene on chromosome 11.1

L-ORD presents with impaired dark adaption and nyctalopia in the 5th–6th decades, progressing to severe sight impairment with loss of central and peripheral vision.2 However, clinical signs preceding visual loss can be overlooked, which leads to delays in diagnosis.3–5 For example, peripupillary iris atrophy and long anteriorly inserted zonules have been reported as early features of L-ORD.6 Additionally, there is variation in clinical presentation and natural history in L-ORD, which can be attributed to different mutations in the C1QTNF5 gene. De Zaeytijd et al demonstrated three phenotypes in patients with L-ORD associated with the c.562C>A p.(Pro188Thr), which differed in terms of clinical signs, age of onset and disease progression.7 Stanton et al have also demonstrated S163R and P188T mutations that can cause L-ORD.8

Although no treatment is currently available for L-ORD, early diagnosis via genetic confirmation allows family screening. In this context, the combination of clinical examination and multimodal imaging is of paramount importance in the diagnosis and monitoring of patients with L-ORD. Moreover, the diagnosis can be challenging as it mimics other retinal degenerative diseases in its early stages, in particular age-related macular degeneration (AMD).9 The first retinal manifestation of L-ORD is generally the onset of yellow-white punctate lesions in the mid-periphery called reticular pseudodrusen (RPD). These are commonly misdiagnosed as drusen.1 2 10 Over time, these deposits are associated with areas of atrophy separated by intervening spaces of intact retinal pigment epithelium (RPE) with scalloped borders.9–11 In the S163R variant, progression of RPE degeneration has been reported from the temporal periphery to the fovea leading to severe central visual loss.7 10 Khan et al have recently described the separation of the Bruch’s membrane (BM) from the RPE on spectral domain-optical coherence tomography (SD-OCT) and correlated this finding to the BM thickening detected histopathologically in patients with L-ORD.12 Macular choroidal neovascular membranes (CNVs) have also been reported as a cause of visual loss in L-ORD.13

A better characterisation of the disease, in particular using repeatable and standardised imaging modalities, is important to assist the diagnosis of L-ORD. This study aims to present peripheral outer retinal (OR) corrugations as a diagnostic finding of L-ORD.

Methods

This retrospective study reviewed consecutive patients diagnosed with L-ORD in the inherited retinal diseases clinic at Sunderland Eye Infirmary, Sunderland, UK, between January 2021 and June 2023. Diagnoses were confirmed via genetic testing for the C1QTNF5 gene and consistent phenotypic findings. Full field electroretintogram was used to confirm a rod-cone dystrophy pattern.14

Demographic and clinical data were extracted from participants’ medical records. Visual acuity was routinely measured using Early Treatment of Diabetic Retinopathy Study (ETDRS) letters. In addition, historic diagnostic investigations were reviewed, including SD-OCT and ultra-widefield (UWF) fundus photography.

Patients had additional imaging to capture the OR corrugations if they were identified on clinical examination. Additional images included: five UWF pseudocolour images and autofluorescence photos using Optos 200Tx (Optos plc, Dunfermline, Scotland, UK), the first image centred on the fovea and one in each peripheral quadrant (nasal, temporal, superior and inferior). True colour fundus photos were obtained using the Canon CX-1 digital retina camera (Canon Inc, Tokyo, Japan). Spectral-domain OCT (Heidelberg Engineering, Heidelberg, Germany) was acquired with a minimum of 19 OCT B-scans in 20° pattern centred on the fovea. The cut-off for the signal strength was set to >15 Q score. The 55° lens on the Spectralis Wide Field Imaging Module was used to acquire horizontal and vertical scans over the retinal areas suspicious for OR corrugations at DFE and/or UWF photos.

Results

Nine patients with confirmed L-ORD were reviewed. Demographic and clinical data are summarised in table 1. All patients had a complete bilateral ophthalmic examination, as described above. Typical findings of LORD, including RPD, temporal atrophic changes and leopard spotting, were detected in all eyes; whereas, OR corrugations were identified in 13 out of 18 eyes (72%). All patients had confirmed mutations in the C1QTNF5 gene. Five patients (56%) were female and mean best-corrected visual acuity was 50. On full-field, ERG, a severe scotopic dysfunction was present in all eyes, while attenuated photopic responses were detected bilaterally in two of eight patients.

On UWF pseudocolour photos, OR corrugations showed a yellowish linear appearance and were more frequently located close to the edges of the areas of chorioretinal atrophy (figures 1–4). On OCT scans, OR corrugations appeared as undulated RPE profile combined with contiguous hyper-reflective ‘bumps’ above BM (figures 1–4). In all eyes with OR corrugations, the temporal quadrants were involved, more commonly the superotemporal (69% of the eyes showing OR corrugations). Furthermore, in patients where OR corrugations were present in both eyes, they appeared in the same location in each eye.

Request permission

Request permission

Request permission

Request permission

Discussion

We present OR corrugations as a diagnostic finding of L-ORD and propose them as an additional common feature of the L-ORD phenotype, identifiable on both clinical examination and imaging, that can aid diagnosis and disease monitoring. In contrast to the L-ORD-associated retinal folds previously described by Khan et al at the posterior pole within areas of established atrophy, we detected OR corrugations in the mid-periphery and far-periphery; close, but not within, the areas of chorioretinal atrophy mainly involving the posterior pole.15

Timely diagnosis of L-ORD has various challenges due to its late-onset nature, its ability to mimic other retinal conditions and the lack of understanding of its disease course.9 10 In particular, it mimics AMD. Clinically, L-ORD affects patients’ peripheral vision, presents at a younger age and has a positive family history.16 In L-ORD, atrophic areas are hypoautofluorescent and give a ‘leopard spot’ fundal appearance on autofluorescent imaging. These autofluorescent changes are more pronounced and widespread than those typically seen in AMD.16

The involvement of RPE and BM in the OR corrugations described in L-ORD is consistent with the aetiopathology of the disease. C1QTNF5 plays an important role in the adhesion between the RPE and BM, and the mutant protein in L-ORD is thought to impair this connection leading to accumulation of subretinal deposits.10 Kuntz et al noted a thick layer of extracellular deposits between BM and RPE in a donor eye with presumed L-ORD.9 In particular, they described two distinct sublayers: the outer layer was rich in lipids and collagen and positive for amyloid P and lysozyme, and the inner layer contained collagen, mucosubstances, elastin, Müller processes and rhodopsin positive rod neurites.9 These bands of deposits underlie areas of profound photoreceptor loss. This is similar to deposits seen in Sorsby’s fundus dystrophy where the band of deposit is thought to impair nutrient transport to the inner retina.2 9 The outer layer extracellular deposits share common constituents with atrophic areas in AMD. Fleckenstein et al described hyperautofluorescent crown-like structures on OCT in areas of geographic atrophy (GA) secondary to AMD.16 Subsequently, Ooto et al coined the term ‘OR corrugations’ referring to a curvilinear undulating sheet of hyper-reflective material above BM within atrophic areas, continuous with the posterior surface of RPE band, extending into neighbouring regions without RPE.17 Through histopathological analysis, these corrugations were attributed to basal laminar deposits consisting of basement membrane proteins embedded with fibrous long-spacing collagen which are initially discrete and then coalesce to form folds.18 These basal laminar deposits were found to split the RPE from its basal lamina in the areas of RPE loss, which corresponded to the curvilinear band seen on OCT images. These deposits accumulate in AMD and form sublayers rich in lipoprotein-derived debris that form the corrugations.18

Bonnet et al described the same features in patients with GA as heterogeneous hyper-reflective internal boundary with a hyporeflective core and called them ‘hyper-reflective pyramidal structures (HPS)’.19 The authors suggested that HPS may be an advanced soft drusen, characterised by the spontaneous drusen regression and a residual scaffold visible on OCT.19 In contrast to the AMD-related OR corrugations, which are found in the posterior pole, in our series L-ORD-associated OR corrugations were identified the peripheral retina, outwith areas of atrophy. This could represent early deposition of basal laminar deposits in L-ORD in areas that were not previously thought to be affected. Thus, OR corrugations might represent coalescence of drusen-like deposits, causing a band to form beneath the RPE that throws the retina into folds. Further longitudinal follow-up of these patients is required to assess whether OR corrugations are associated with atrophy in L-ORD. Ooto et al observed similar corrugations corresponding to areas of neovascularisation in AMD, but it is our understanding that this is not the case for our cohort.17 Keenan et al reported massive advancing non-exudative type 1 CNV in the mid-peripheral fundus with CTRP5-LORD and suggest that hypoperfusion of the choroid may precede neovascularisation and that non-exudative neovascularisation can be protective against atrophy.20 We did not find any association with retinal exudative or non-exudative neovascularisation in our cohort; however, our knowledge of this is limited by historic imaging focusing only on the posterior pole. As eccentric OCTs are not routinely performed we cannot exclude the possibility of eccentric CNV preceding the development OR corrugations and further studies and increased use of fundal angiography at diagnosis may be beneficial to further examine this possibility. Given the peripheral location of the OR corrugations, performing OCT angiography would be technically challenging.

We suspect that the OR corrugations may precede the development of atrophy. It will be interesting to observe this cohort longitudinally to see how these findings change over a long-term follow-up.

This study highlighted the paramount importance of multimodal imaging in the assessment of patients with L-ORD.21 We suggest that the identification of OR corrugations may help in the diagnosis and characterisation of L-ORD.